Antibodies specific for carcinogen-DNA adducts have probed the nature, extent, and consequences of in vitro and in vivo DNA modification. Biological samples substituted with 2- acetylaminofluorene (AAF), cis-diamminedichloroplatinum II (cis- DDP) and benzo(a)pyrene (BP) have been analyzed by quantitative immunoassays and by immunohistochemical procedures developed to localize adducts in situ. In hepatic DNA of rats chronically fed a carcinogenic dose of AAF, adduct removal has biphasic, suggesting two genomic compartments - one from which adducts are removed rapidly and another from which they are removed slowly. These kinetics are not due to different liver cell types or DNA associated with more or less tightly bound chromatin regions including the nuclear matrix. Comparison of adduct and tumor dose responses in bladder and liver of mice fed AAF chronically showed linear adduct and liver tumor formation, but non-linear bladder tumor formation. Immunohistochemical localization of 2- aminofluorene-DNA (AF-DNA) adducts in livers of rats fed AAF demonstrated a non-uniform adduct distribution and no adducts in preneoplastic foci induced by several different hepatocarcino- genesis protocols. Cis-DDP-DNA adducts were measured in DNA from nucleated peripheral blood cells and in several tissues obtained from cancer patients at multiple times during courses of cis-DDP therapy and/or at autopsy. Adduct accumulation occurred as a function of total cumulative dose. Disease response data on 55 ovarian cancer Patients and 17 testicular cancer patients indicated that individuals with high adduct levels have a high rate of complete response to therapy. BP-DNA antigenicity was measured in enzyme-linked immunosorbent assays (ELISA) blood cell DNA of 43 firefighters and 40 controls. In both groups about 30% of the samples were positive, but mean adduct levels were higher in samples from firefighters. Other factors which contributed to sample positivity were the ingestion of charcoal broiled foods and smoking.